Research ArticleBIOCHEMISTRY

A tail of two voltages: Proteomic comparison of the three electric organs of the electric eel

See allHide authors and affiliations

Science Advances  05 Jul 2017:
Vol. 3, no. 7, e1700523
DOI: 10.1126/sciadv.1700523
  • Fig. 1 Locations of phosphorylated residues that were detected in representative members of four important electrogenic protein families.

    (A) Illustration of electrogenic proteins in electric organ discharge. Action potential is driven by massive Na+ influx mediated by Na+ channels (AChRs and voltage-gated Na+ channels) in the innervated membrane face, whereas the Na+/K+-ATPase works to maintain the negative membrane potentials at the noninnervated face. This results in a large transcellular potential difference, and because electrocytes are arranged massively in series, like batteries in a flashlight, voltages summate. ADP, adenosine 5′-diphosphate; KiR, inwardly rectifying potassium channel; NaV, voltage-gated sodium channel. (B) Relative location of identified phosphorylated residues in electrogenic proteins. Green lines indicate the location of phosphorylated residues identified in these data that have been described in mammals (9). Red lines indicate novel phosphorylated residues in E. electricus. Blue indicates phosphosites identified in E. electricus that fall within less than five residues of a known phosphosite in mammals. Amino acid (aa) length given is for gene model chosen for analysis (see table S1 for gene model selection, when more than one was possible, if applicable). (C) Phosphorylated residues (yellow or pink) of Na+/K+-ATPase ATP1a2a from E. electrophorus are shown in their homologous positions in the crystal structure of the α subunit of Na+/K+-ATPase from Sus scrofa (PDB accession 3WGU). Ribbon colors: red, nucleotide-binding domain; purple, phosphorylation domain; blue, actuator domain; green, membrane domain. (D) Phosphorylated residues (red or cyan) of voltage-gated Na+ channel SCN4aa are shown in a transmembrane topology cartoon that is predicted for this E. electrophorus protein (UniProt accession P02719), in lieu of a high-resolution three-dimensional structure that is not yet available. (E) Phosphorylated residues (red or cyan) of the nicotinic AChR β subunit CHRNB1 from E. electrophorus are shown in a transmembrane topology cartoon that is predicted for this E. electrophorus protein based on its homology with mouse CHRNB1 (UniProt accession P09690), also in lieu of a high-resolution three-dimensional structure. The phosphorylated residues are found intracellularly between transmembrane domains M3 and M4, a region whose structure is currently undefined in high-resolution structures of nicotinic AChRs. (F) Phosphorylated residues (yellow) of ACHE from E. electrophorus are shown in their homologous positions in the crystal structure of ACHE from T. californica (PDB accession 3I6M). Color code in (C) to (F) for phosphorylated residues: yellow or red, phosphorylation at these residues is not found in mammals and is potentially relevant to unique activities of electrocytes; pink or cyan, phosphorylation at these residues is found in mammals.

  • Fig. 2 Proteins important for electric organ discharge in E. electricus show distinct patterns of abundance across electric organs.

    (A) Relative abundance of electrogenic ion transporters in electric organs and muscle. The general pattern observed among most ion transporters (main > Hunter’s > Sachs’) is broken by a voltage-gated potassium channel. (B) Total abundance of potassium channels detected in this study. Values shown are log2-normalized summed reporter ion intensity values in table S2.

  • Fig. 3 Clustering reveals relatedness among three distinct electric organs.

    (A and B) Clustering of biological replicate tissues by (A) relative protein abundance and (B) relative phosphopeptide abundance. Quantitative values for protein groups and phosphopeptides were normalized as indicated in Materials and Methods. Clustering was performed using complete-linkage hierarchical clustering, and the numbers inside each square indicate Euclidean distance measurement, with darker colors indicating small distances. The data indicate that the identical tissues from two individual E. electricus specimen cluster together (that is, main electric organ from eel 1 clusters with main electric organ from eel 2), and muscle clusters separately from the three electric organs, as expected in both. The heat map suggests that because Hunter’s electric organ and Sachs’ electric organ cluster most closely to one another, they are more similar to one another and are more distinct from main electric organ. (C and D) Venn diagrams showing overlap of top 10% most abundant (C) proteins and (D) phosphopeptides in each tissue and fish. Venn diagrams reveal that Hunter’s electric organ has the least unique proteins and phosphopeptides compared to main electric organ, Sachs’ electric organ, and muscle. Protein-level Venn diagrams (C) show that, at a protein level, the largest shared group is among proteins most abundant in all four tissues tested (muscle and electric organ). Phosphopeptide-level Venn diagrams (D) indicate that at a phosphopeptide abundance level, the largest shared group is among the three electric organs, indicating that there is distinct protein phosphorylation in the electric organs relative to muscle. (E and F) k-means clustering depicts co-regulation in specific electric organ tissues at the protein level (E) and phosphopeptide level (F). Mean log2 (sample/median) abundance values for the tissues in each cluster are indicated in each box. Darker blue colors indicate more negative values, whereas darker red colors indicate more positive values. Results indicate that there are electric organ–specific patterns of protein and phosphopeptide abundance.

  • Fig. 4 Phosphopeptides that differ in abundance compared to protein abundance.

    Bar graphs showing the relative abundance of phosphopeptides normalized to protein abundance across tissues. NA, the phosphopeptide was not detected in a sample.

  • Fig. 5 Correlation of RNA expression and protein abundance.

    (A). Plots showing the correlation of RNA expression and protein abundances for all proteins detected in unenriched proteomics experiment. See the “Correlation of RNA expression and protein abundance values” section for details on how graphs were generated. DE, differentially expressed genes or proteins. (B) Table showing counts of gene models falling into each category depicted on graphs in (A).

  • Fig. 6 Overview of experimental method.

    Flowchart showing overview of procedures in this study.

  • start11M 11e+100RM 11E 11W 11P 11e3
    stop11M 11e+100RM 11E 11W 11P 11e3
    tss11M 11e+100RM 11E 11W 11P 11
    tts11M 11e+100RM 11E 11W 11P 11
    ass11M 11e+100RM 11E 11W 11P 1100
    dss11M 11e+100RM 11E 11W 11P 1100
    exonpart1.997M 11e+100RM 11E 11e2W 11.007P 11
    exon11M 11e+100RM 11E 11e4W 11P 11e4
    intronpart11M 11e+100RM 11E 11W 11P 11
    intron1.3M 11e+100RM 11E 11e6W 11P 1100
    CDSpart10.985M 11e+100RM 11E 11W 11P 11e5
    CDS11M 11e+100RM 11E 11W 11P 11
    UTRpart1.96M 11e+100RM 11E 11W 11P 11
    UTR11M 11e+100RM 11E 11W 11P 11
    irpart11M 11e+100RM 11E 11W 11P 11
    nonexonpart11M 11e+100RM 11.01E 11W 11P 11
    genicpart11M 11e+100RM 11E 11W 11P 11

Supplementary Materials

  • Supplementary material for this article is available at http://advances.sciencemag.org/cgi/content/full/3/7/e1700523/DC1

    fig. S1. Phosphorylation sites in the C-terminal domain of E. electricus SCN4aa.

    fig. S2. Normalization of proteomic and phosphoproteomic data.

    table S1. Novel and known phosphosites in E. electricus proteins.

    table S2. Median-normalized channel intensity values for unenriched, whole-proteome samples.

    table S3. Normalized intensity ratios, log2 (tissue/median), on a per-peptide basis.

    table S4. Potential roles for nonelectrogenic proteins and phosphopeptides differentially abundant in electric organs.

    table S5. Expression values (RNA) for all predicted genes in assembly.

    table S6. Intensity ratios, log2 (tissue/median), on a per-protein group basis.

    table S7. Phosphopeptides in electric organ discharge–related proteins that differ in abundance compared to protein abundance.

    table S8. Raw output from Proteome Discoverer, unenriched, whole-proteome samples.

    table S9. Raw output from Proteome Discoverer, titanium dioxide–enriched phosphopeptides.

    table S10. Correlation of RNA and protein abundance values.

    table S11. Comparison of new genome assembly and gene annotations to the previous assembly.

    table S12. Median-normalized channel intensity values for titanium dioxide–enriched phosphopeptides.

    table S13. Peptide and protein group counts for unenriched, whole-proteome experiment, and TiOX-enriched phosphoproteome experiment.

  • Supplementary Materials

    This PDF file includes:

    • fig. S1. Phosphorylation sites in the C-terminal domain of E. electricus SCN4aa.
    • fig. S2. Normalization of proteomic and phosphoproteomic data.
    • Legends for tables S1 to S10
    • table S11. Comparison of new genome assembly and gene annotations to the previous assembly.
    • Legend for table S12
    • table S13. Peptide and protein group counts for unenriched, whole-proteome experiment, and TiOX-enriched phosphoproteome experiment.

    Download PDF

    Other Supplementary Material for this manuscript includes the following:

    • table S1 (Microsoft Excel format). Novel and known phosphosites in E. electricus proteins.
    • table S2 (Microsoft Excel format). Median-normalized channel intensity values for unenriched, whole-proteome samples.
    • table S3 (Microsoft Excel format). Normalized intensity ratios, log2 (tissue/median), on a per-peptide basis.
    • table S4 (Microsoft Excel format). Potential roles for nonelectrogenic proteins and phosphopeptides differentially abundant in electric organs.
    • table S5 (Microsoft Excel format). Expression values (RNA) for all predicted genes in assembly.
    • table S6 (Microsoft Excel format). Intensity ratios, log2 (tissue/median), on a perprotein group basis.
    • table S7 (Microsoft Excel format). Phosphopeptides in electric organ discharge–related proteins that differ in abundance compared to protein abundance.
    • table S8 (Microsoft Excel format). Raw output from Proteome Discoverer, unenriched, whole-proteome samples.
    • table S9 (Microsoft Excel format). Raw output from Proteome Discoverer, titanium dioxide–enriched phosphopeptides.
    • table S10 (Microsoft Excel format). Correlation of RNA and protein abundance values.
    • table S12 (Microsoft Excel format). Median-normalized channel intensity values for titanium dioxide–enriched phosphopeptides.

    Download Tables S1 to S12

    Files in this Data Supplement: